Technology companies are getting closer to creating computers that are modeled on the human brain, after decades of disappointing efforts. Chipmakers such as Qualcomm, IBM, and Intel are now merging their knowledge of microprocessors and neuroscience to develop a new brain-inspired products. For example, Qualcomm next year plans to release its Zeroth chip, based on a type of processor called a neural-inspired processing unit (NPU). Although an NPU is still a computer chip made of silicon and patterned with transistors, it can perform qualitative functions. Theoretically, an NPU based on mathematical modeling and human biology could learn like a human, and Qualcomm hopes that in the next few years NPUs will appear in devices ranging from smartphones to servers in a cloud architecture. IBM is working on a cognitive computing paradigm called TrueNorth, which attempts to model computing on the brain and could alter computer vision, motion detection, and pattern matching. TrueNorth can be programmed with "corelets" that teach the computer different functions. Meanwhile, Intel's neuromorphic chip architecture is based on devices that use the spinning of an electron to perform a variety of computing functions. Intel says neuromorphic hardware based on spin devices can perform analog data-sensing, data conversion, cognitive computing, associative memory, programmable logic, and analog and digital-signal processing.